# Explanation Generation on GDELT (test) Source: [Explanation Generation on GDELT (test)]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:37:18.659Z Evaluation of model performance on generating textual explanations using the test split of the GDELT dataset, measured using standard NLP generation metrics like BLEU, ROUGE, and BERTScore. A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. | Scope | Recorded value | | --- | --- | | Task | Explanation Generation | | Dataset (including any stated split) | GDELT (test) | | Metric | BLEU-4 | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 17 of 17 | | Coverage | All results currently returned for this benchmark | Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol. ## Metrics - [BLEU-4](): Higher is better; unit: PERCENT. - [ROUGE-L](): Higher is better; unit: PERCENT. - [METEOR](): Higher is better; unit: PERCENT. - [BERTScore (F1)](): Higher is better; unit: PERCENT. ## Recorded results | Method | BLEU-4 | Result date | Report paper | | --- | --- | --- | --- | | GETER | 34.46 | Not recorded | 68fa78a61e209131cea799af | | Llama3-8B-Instruct | 34.32 | Not recorded | 68fa78a61e209131cea799af | | GETER | 33.81 | Not recorded | 68fa78a61e209131cea799af | | GETER | 32.18 | Not recorded | 68fa78a61e209131cea799af | | Mistral-7B-Instruct | 30.93 | Not recorded | 68fa78a61e209131cea799af | | Qwen2.5-7B-Instruct | 26.1 | Not recorded | 68fa78a61e209131cea799af | | Mistral-7B-Instruct | 18.22 | Not recorded | 68fa78a61e209131cea799af | | Llama3-8B-Instruct | 18.12 | Not recorded | 68fa78a61e209131cea799af | | Qwen2.5-7B-Instruct | 17.15 | Not recorded | 68fa78a61e209131cea799af | | GPT-4o | 9.16 | Not recorded | 68fa78a61e209131cea799af | | Qwen2.5-7B-Instruct | 7.55 | Not recorded | 68fa78a61e209131cea799af | | Mistral-7B-Instruct | 7.46 | Not recorded | 68fa78a61e209131cea799af | | GPT-4o | 5.95 | Not recorded | 68fa78a61e209131cea799af | | Llama3-8B-Instruct | 5.61 | Not recorded | 68fa78a61e209131cea799af | | Mistral-7B-Instruct | 4.89 | Not recorded | 68fa78a61e209131cea799af | | Qwen2.5-7B-Instruct | 3.76 | Not recorded | 68fa78a61e209131cea799af | | Llama3-8B-Instruct | 2.38 | Not recorded | 68fa78a61e209131cea799af | ### GETER BLEU-4: 34.46. Result ID: 699598913badfaffe693d0eb. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"backbone":"Llama3-8B-Instruct"}. This record does not assert independent reproduction. ### Llama3-8B-Instruct BLEU-4: 34.32. Result ID: 699598913badfaffe693d0ea. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"training\_protocol":"LoRA","reasoning\_chains":true}. This record does not assert independent reproduction. ### GETER BLEU-4: 33.81. Result ID: 699598913badfaffe693d0f0. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"backbone":"Qwen2.5-7B-Instruct"}. This record does not assert independent reproduction. ### GETER BLEU-4: 32.18. Result ID: 699598913badfaffe693d0f5. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"backbone":"Mistral-7B-Instruct"}. This record does not assert independent reproduction. ### Mistral-7B-Instruct BLEU-4: 30.93. Result ID: 699598913badfaffe693d0f4. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"training\_protocol":"LoRA","reasoning\_chains":true}. This record does not assert independent reproduction. ### Qwen2.5-7B-Instruct BLEU-4: 26.1. Result ID: 699598913badfaffe693d0ef. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"training\_protocol":"LoRA","reasoning\_chains":true}. This record does not assert independent reproduction. ### Mistral-7B-Instruct BLEU-4: 18.22. Result ID: 699598913badfaffe693d0f3. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"training\_protocol":"LoRA","reasoning\_chains":false}. This record does not assert independent reproduction. ### Llama3-8B-Instruct BLEU-4: 18.12. Result ID: 699598913badfaffe693d0e9. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"training\_protocol":"LoRA","reasoning\_chains":false}. This record does not assert independent reproduction. ### Qwen2.5-7B-Instruct BLEU-4: 17.15. Result ID: 699598913badfaffe693d0ee. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"training\_protocol":"LoRA","reasoning\_chains":false}. This record does not assert independent reproduction. ### GPT-4o BLEU-4: 9.16. Result ID: 699598913badfaffe693d0e6. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"training\_protocol":"zero-shot","reasoning\_chains":true}. This record does not assert independent reproduction. ### Qwen2.5-7B-Instruct BLEU-4: 7.55. Result ID: 699598913badfaffe693d0ed. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"training\_protocol":"zero-shot","reasoning\_chains":true}. This record does not assert independent reproduction. ### Mistral-7B-Instruct BLEU-4: 7.46. Result ID: 699598913badfaffe693d0f2. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"training\_protocol":"zero-shot","reasoning\_chains":true}. This record does not assert independent reproduction. ### GPT-4o BLEU-4: 5.95. Result ID: 699598913badfaffe693d0e5. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"training\_protocol":"zero-shot","reasoning\_chains":false}. This record does not assert independent reproduction. ### Llama3-8B-Instruct BLEU-4: 5.61. Result ID: 699598913badfaffe693d0e8. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"training\_protocol":"zero-shot","reasoning\_chains":true}. This record does not assert independent reproduction. ### Mistral-7B-Instruct BLEU-4: 4.89. Result ID: 699598913badfaffe693d0f1. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"training\_protocol":"zero-shot","reasoning\_chains":false}. This record does not assert independent reproduction. ### Qwen2.5-7B-Instruct BLEU-4: 3.76. Result ID: 699598913badfaffe693d0ec. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"training\_protocol":"zero-shot","reasoning\_chains":false}. This record does not assert independent reproduction. ### Llama3-8B-Instruct BLEU-4: 2.38. Result ID: 699598913badfaffe693d0e7. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"training\_protocol":"zero-shot","reasoning\_chains":false}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.